The eye needs to biosynthesize 11-cis-retinoids because the chromophore of rhodopsin is 11-cis-retinal. The critical metabolic step is the endergonic isomerization of free all-trans-retinol (vitamin A) into 11-cis-retinol. This isomerization process can take place in isolated membranes from the retinal pigment epithelium in the absence of added energy sources. Specific binding proteins probably do not serve as an energy source, and since all of the reactions in the visual cycle are shown here to be reversible, trapping reactions also do not participate in the isomerization reaction. One previously unexplored possibility is that the chemical energy in the bonds of the membrane itself may drive the isomerization reaction. A group transfer reaction is proposed that forms a retinyl ester from a lipid acyl donor and vitamin A. This transfer can drive the isomerization reaction because the all-trans-retinyl ester is isomerized directly to 11-cis-retinol. Thus, the free energy of hydrolysis of the ester is coupled to the thermodynamically uphill trans to cis isomerization. The prediction of an obligate C-O bond cleavage in the vitamin A moiety during isomerization is borne out. Although the natural substrate for isomerization is not known, all-trans-retinyl palmitate is processed in vitro to 11-cis-retinol by pigment epithelial membranes.

The key biochemical process of the vertebrate visual cycle required for rhodopsin regeneration, 11-cisretinoid production from all-trans-retinoids, is shown to occur in vito. A 600 x g supernatant from a frog retina/pigment epithelium homogenate transforms added all-trans-[3H]retinol, in a time-dependent fashion, to a mixture of 11-cis-retinol, 11-cis-retinal, and 11-cis-retinyl palmitate. 13-cis-Retinoids are formed in only minor amounts by nonspecific processes. Studies using washed particulate fractions of the 600 x g supernatant indicate that all-trans-[3H]retinol is isomerized to 11-cis-retinoids much more effectively than is all-trans-[3H]retinal or all-trans-[3H]retinyl palmitate. The li-cis-retinoid biosynthetic activity is heat-labile, sedinmentable by highspeed centrifugation, and largely found in the pigment epithelium rather than in the neural retina.The absorption of light by rhodopsin in the vertebrate eye results in the cis-to-trans photoisomerization of its 11-cisretinal chromophore, bound as a protonated Schiff base to lysine, eventually leading to the hydrolysis and release of the all-trans-retinal (1,2). One of the rhodopsin conformers during this bleaching process, metarhodopsin II, catalyzes the binding of GTP in exchange for GDP by a retinal GTP-binding (G) protein, thus initiating the process of visual transduction (3, 4). Under bright light conditions the alltrans-retinal liberated by bleaching is reduced, esterified to long-chain fatty acids, and stored in the pigment epithelium of the eye (5). When a light-adapted animal encounters a dark environment, 11-cis-retinoids must be regenerated from the stores of all-trans-retinoids, and the 11-cis-retinal produced can then combine with opsin to form rhodopsin. The ocular biosynthesis of 11-cis-retinal in the dark is, by definition, a thermal process. It is also an endergonic process because at thermal equilibrium 11-cis-retinoids represent only 0.1% of the retinoids (6), while in a dark-adapted eye at least 75% of the retinoids are in the 11-cis form, with the remaining retinoids in the all-trans form (5). The driving force cannot simply arise from the stereospecific combination of 11-cisretinal with opsin because in many higher vertebrates such as man and amphibians, there is a 2-to 3-fold excess of retinoids over opsin in the eye (5, 7).A particularly vexing problem in visual science has been the mechanism of 11-cis-retinoid biosynthesis in the eye. Major hurdles include the identification of the substrate for isomerization (retinol, retinal, or retinyl ester), the nature of the energy source that drives the biosynthesis of 11-cisretinoids, and most importantly, the identification of the isomerizing system capable of producing 11-cis-retinoids in vivo. No system that can produce 11-cis-retinoids in vitro in darkness has yet been confirmed (8). Over the years several attempts have been made at the identification of an isomerase enzyme specific for retinal (9, 10); however, these attempts have suffered from a misidentification of rho...

The biosynthesis of 11-cis-retinol in the retinal pigment epithelium requires two consecutive enzymatic reactions. The first involves the esterification of all-trans-retinol by lecithin retinol acyltransferase (LRAT). The second reaction involves the direct conversion of an all-trans-retinyl ester into 11-cis-retinol by an isomerase-like enzyme. This latter reaction couples the free energy of hydrolysis of an ester to the thermodynamically uphill trans to cis conversion, thus providing the energy to drive the latter process. In this paper both enzymes are studied with respect to their substrate specificities to provide information on mechanism. The isomerase is shown to be highly specific with respect to the ionylidene ring system and substitution at C15, whereas sterically bulkier substituents at C9 and C11 are permitted. C5 and C13 demethyl retinoids are isomerized, removing from consideration isomerization mechanisms involving C-H abstraction at the C5 or C13 methyl groups of the retinoid. On the other hand, C9 demethyl retinoids are not isomerized. A C-H abstraction mechanism is unlikely at the C9 methyl group as well, because no kinetic deuterium isotope effect is found with all-trans-19,19,19-trideuterioretinoids and isomerization of unlabeled retinoids occurs without the incorporation of deuterium when the isomerization is performed in D2O. LRAT proved to be broadly specific for retinols but was relatively inert with other hydrophobic alcohols including cholesterol. The enzyme is also highly specific for phosphatidylcholine analogues versus other potential membranous acyl donors such as phosphatidylethanolamine and phosphatidylserine.

No abstract

No abstract